trainer_state.json

{
  "best_metric": 0.8810666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0403/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 2.85232922088503e-05,
      "loss": 2.1739,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7712,
      "eval_loss": 1.01809823513031,
      "eval_runtime": 6.6948,
      "eval_samples_per_second": 560.135,
      "eval_steps_per_second": 8.813,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.0539660453796387,
      "learning_rate": 2.4368144166557088e-05,
      "loss": 0.6532,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8330666666666666,
      "eval_loss": 0.6131467223167419,
      "eval_runtime": 7.5386,
      "eval_samples_per_second": 497.438,
      "eval_steps_per_second": 7.826,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 18.31244659423828,
      "learning_rate": 1.8357520279328874e-05,
      "loss": 0.3232,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.48682910203933716,
      "eval_runtime": 6.8227,
      "eval_samples_per_second": 549.64,
      "eval_steps_per_second": 8.648,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 1.1701616683386336e-05,
      "loss": 0.1688,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8730666666666667,
      "eval_loss": 0.43521997332572937,
      "eval_runtime": 6.182,
      "eval_samples_per_second": 606.597,
      "eval_steps_per_second": 9.544,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.409283638000488,
      "learning_rate": 5.679298183982636e-06,
      "loss": 0.088,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.4245626926422119,
      "eval_runtime": 6.1528,
      "eval_samples_per_second": 609.477,
      "eval_steps_per_second": 9.589,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 5.281814098358154,
      "learning_rate": 1.5030589901472829e-06,
      "loss": 0.0478,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.4164716899394989,
      "eval_runtime": 6.1543,
      "eval_samples_per_second": 609.329,
      "eval_steps_per_second": 9.587,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}